Crop Phenology Studies Using RGB Drone Data

The drone is used in precession agriculture for spraying pesticides, phenology study, and crop classification. Crop phenology is an essential aspect of agricultural resource management and the classification of various crops. Any automated model should be able to identify or classify phenology stages solely on the statistical feature's assessment rather than the visual inspection. The crop has different colors during the growth stages; therefore, the color-based model is attempted for phenology studies with the help of drone RGB images. Rice, soybean, sugarcane, and wheat crops are selected for the phenology study. Various color-based features are used for phenology studies, and it is found that a* band of la*b color space model and 2*G-R-B (Ex-green) bands can classify the crop growth and harvesting stage. The ripening stage can be identified using the R-B band combination. Therefore, these three features combination can be used to identify different growth stages such as stem-elongation, ripening, harvested, and non-harvested stage.